In Tiger I think it's just in the System Preferences. It's very trivial to swap Command and Option, for example, for dealing with windows keyboards. More complicated things likely require messing with the keyboard maps, which is not trivial under linux either.
The option for turning on keyboard navigation dialog boxes and stuff (that I mentioned earilier) is in the System Preferences under "Keyboard and Mouse" under "Keyboard Shortcuts." Just in case anyone is interested in that.
Will clips from shows like the Simpsons and the Family Guy start disappearing from youtube? I believe they are legal due to fair use. But we all know how copyright holders feel about that these days.
OS X is one of the few OS's to allow total customization of keyboard short-cuts. You can assign almost any combination to any menu item on any program. The downside is you have to do it from System Preferences. Although MacOS has always been mouse centric, it's actually more keyboard-friendly than windows or even linux (Gnome is only now getting good keyboard shortcut access via atk and other accessibility things). Shortcuts are very consistent and work in almost every program. Command-Q to quit, Command-W to close the window, Command-H to hide the entire app (very useful -- almost eliminates the need for multiple desktops when combined with expose and command-tab), Command-S to save, command-O to open, etc.
One thing that annoyed me to no end was the apparent lack of a way to communicate with dialog boxes using only the keyboard. Most of the time command-first letter works, but often it doesn't. I found that if I turn on some of the accessibility options in system preferences, suddenly I can tab between buttons and use the space bar to activate buttons (enter always activates the default button, not the one you're highlighting).
Knowing about how to set shortcuts, the default shortcuts, and the accessibility options has really made OS X more efficient on the keyboard for me than any other OS (well almost -- I still like activating menus on linux and windows with alt-letter). Certainly it's not as bad you illustrate.
I agree that all GUIs are lousy to a degree. Case in point is CAD software. The old autocad shortcuts (still available on autocad to this day) are the way to fly. Puck in one hand, 2 and 3 letter shortcuts in the other. Modern GUIs just don't lend themselves well to CAD.
Unfortunately I've seen a marked decrease in the effectiveness of grey-listing lately, which is similar in intent to your ideas. What I'm finding is that a lot of spam is now coming from RFC-compliant mail servers. Stock spams in particular always come through after faithfully waiting out the greylist timeout. So obviously some spammers are able to wait, even up to 45 minutes, to send their spam to me. So despite your arguments spammers will find a way to still economically spam while tolerating delays, holding connections open, etc.
From what I can tell, you're in the minority. Most users prefer to have the close button on the active tab, and many prefer to have a close button on every tab. I certainly do. I can easily close off tabs I don't want anymore without have to click on each one to raise it to the top first. Also, having a close button on each tab only really works if the tabs are always the same size. That way you can close many tabs just by clicking in the same area. Before with tab size changing this was impossible because the close button (which I had on with an extension) would always change position.
I don't mind his posts so much some of the time (I never go to his site ever), but he uses dishonest means to drive traffic to his own sites and blogs to get ad revenue. This time he says "You'll find more details and pictures about these mapping robots at ZDNet" but really it's not a ZDnet article at all; it's his blog. Slashdot editors, please either edit his posts to report the truth, or don't post his stories at all.
So why would anyone want to decide right now if they would use the GPLv3 or not? Most projects that are licensed under the GPLv2 have the "version 2 or any later version" clause in them, so developers could, when the GPLv3 is finalized, choose to "fork" their project to that license, or keep it the same.
Discussing the good and bad points of the current GPLv3 draft is valid and we should be doing that here. But to ask the question "should developers switch to it?" is immature and a little silly at this point in time. We could say "given the current draft of the GPLv3, I would not use it for reasons x,y, and z," but that is the extent of it I think.
At this moment in time, if the GPLv3 were actually released, I would probably still use the GPL v2 until I had time to really understand the v3 license, and the things it might mean to my project. Currently I think the GPLv2 has some definite weaknesses that the v3 is trying to address. For example, if I write a nice python library or module under the terms of the GPL (is that possible? I don't know offhand what license that would require), a person can just embed python in a closed-source C program, load my GPL module and use it as an integrated part of his closed-source program. The GPLv2 addresses dynamic linking from the pov of the compiler, the linker, but it doesn't take into account these other use cases that are now common. The spirit of the GPL certain forbids what I have described, unless the GPL applies to the program as well, and maybe the language does too, but it's ambiguous and would likely require a court challenge to decide. This is the type of problem the GPLv3 is trying to solve.
I hope that most GPLv2 people's objections to the v3 license will be addressed in some way. We shall see.
Yes, very good points. So the argument is about whether or not spending money in the space program brings about more economic productivity overall. I believe it can. Especially now that certain parts of the tech industry are going off-shore. Despite the consumer electronics production boom overseas (where the majority of our consumer spending seems to end up), high-tech aviation technology is still largely an American thing. Thus spending money on space and aerospace is a good thing for America, in my opinion.
With the exception of the problem of disparity and wealth distribution, everything I have said is true. Spending one dollar in the economy really does generate several times that in effect. It's similar to depositing a dollar in the bank. Due to how money flow works, one dollar in the bank increases the total money supply in an economy by three times that (depending on the reserve rate). The economy as a whole is similar to a bank, in that money is in use in several places at once. I cannot say what that increase factor is (likely it is smaller than 3 times), but logically it does exist.
Spending money on the space program would have the same effect on the economy as using fiscal policy to control inflation. Government spending in the economy does have a very real and immediate effect on the overall economy. Of course this must be done very carefully so as not to cause overheating, inflation and then later recession. As others have correctly pointed out, there is a problem of national internal debt. But if you take an extreme point of view (I do not espouse this position), paying down the debt has very little positive short-term effect on the economy, and so the debt should just be ignored (after all the money system of profits, debt, debits, credits is all really a figment of our imagination -- an immaterial house of cards).
Of course you could back up your claim with sound economic theory that refutes my statements. I would accept that.
Wasting billions of dollars? You have a strange idea of where money goes. Those billions of dollars that if spent on a space race would ultimately increase the size of our economy (national and global) by up to three fold (if I remember my Econ 101 class correctly). That amplifies the affect of the money and really allows that money to now benefit many more people, that it would if we spent it directly on, say, some kind of welfare assistance. Of course it's the disparity that we should worry about. But still. It's not like our money goes into space with the rockets.
VMWare Player. Add in a text editor and Qemu to create disk images and you've got a virtual machine. Cheap and very effective. Or just use vmware-server (free edition) to create disk images.
I highly doubt there would be any vibration at all. In order to rotate at such high speeds, the rotors need to be perfectly balanced. I don't think the noise level would be too great either. The high rpms and perfect balance would likely mean a very high-frequency noise, one you're unlikely to hear (maybe it would beach whales and make cats and dogs suicidal, though). As for fumes, I don't think you're likely to notice any smell other than the normal out-gassing your plastics and electronics already do (mmm smell that new PowerBook!).
Unfortunately battery technology in and of itself is getting close to the theoretical limit. Lithium is the most reactive metal we can use right now (unless you want to use a radioactive metal). Energy density is still slowly increasing, but this comes at the expense of longer charge times. Right now I admit Lithium technology is pretty good. Out of a 5 cell LiPoly pack, for example you can get up to 45 continuous amps for maybe 8-10 minutes. Enough to have a lot of fun at the RC flying field with a 5 pound plane. But then it's home to charge for several hours.
To answer your question, though. If the price was competitive with current battery technologies, and the energy produced as as good or greater, I'd definitely convert. In fact if the amp draw was possible, I'd stick them in my R/C airplanes. Cleaner than glow power, faster recharge cycles than electric. Oh and I wouldn't mind it in my cell phone either.
Except that we simply don't have the generating or distribution capacity to handle wide-spread electric vehicles. So we can't just use existing power lines, unfortunately. And in the short-term, one study in Canada showed that converting an entire province to electric cars would increase air pollution dramatically because of the reliance on coal-fire electrical plants (of which many more would have to be built). Now ideally, you can clean up electrical generation facilities more easily than individual cars.
Someone on slashdot once calculated the total number of megawatts we'd need to generate and deliver to power electric cars if everyone had them, and it was staggering, far beyond our current capacity (no pun intended).
The "Hyrdogen Economy" will never exist. It has not been and never will be an energy source. Energy storage maybe, but not a source. Even if you can take cleanly generated electricity and make hydrogen via electrolosis, I don't think it's viable, except in small applications like laptop fuel cells. Today almost all hydrogen comes from natural gas anyway. I guess these guys have just developed a more efficient way to get get the hydrogen extracted from it, but the byproducts are still the same (carbon dioxide gas).
In the long run, I think it is better to work on a carbon-neutral way of generating more complex hydrocarbons, such as bio-diesel, which will not release any greenhouse gases, yet have a very high energy density that we need. I believe it is harmful to be releasing so much net carbon dioxide into the atmosphere and the oceans.
Right. This is about software distribution, not software use by an end-user. The GPL is *not* an EULA; it's a license to distribute the program. The GPL explicitly states that you do not have to agree to the terms of the GPL in order to use the program.
I suppose that by these same arguments you could argue that EULA's must be valid because nothing else grants you the right to use copyrighted materials when you buy a program from a closed-source vendor. But EULAs go beyond copyrights into contract law, I believe they are a different beast than source code licensing.
Actually it appears that in the absence of dust storms, the color of the sky on Mars is, in fact, blue, much like earth's. In some pictures I've seen, you might mistake the martian landscape for Arizona!
Although take it all with a grain of salt. There are good arguments for and against the blue sky. The JPL, unfortunately has never put forth a lot of convincing proof for the red sky, fueling this debate further.
Sending rockets out into space is pretty easy, but the real trick is orbit. Cheap shots to the upper atmosphere don't do a lot of good in terms of launching satellites and other objects into orbit, although I'm sure they can provide experience with the technology. Achieving orbit requires a lot more energy. There's a reason missiles and rockets are the size they are.
Tango is just an icon set, done with SVG. They use an elipse as a shadow instead of a "real" shadow. Big deal. The icons still are a cut above XP's icon set. I dont't know much about the workings of SVG, but I'd say real shadows would be difficult but doable. In any case, it's just an icon! It's not even 3D. At 128x128 it's not going to matter that much. I think you're just being pedantic. You're welcome to submit better icon shadows to the project rather than complain about things being unprofessional. Your professional help would be greatly appreciated if you feel inclined.
Re:Ten years for a DOS clone..? You got to be joki
on
FreeDOS 1.0 Released
·
· Score: 1
Not sure why the parent comment was moderated insightful. FreeDOS has a very bright future and will continue to be used in its niche areas. So indeed it is a very useful project. So I am very grateful for the hard work of the freeDOS developers over all these years. Just the other day I used FreeDOS under dosemu to resurrect some old PowerBASIC code.
After a couple of training messages, dspam is picking these spams (with or without the animated gif) with great accuracy. Even if dspam has no idea what the spam message is saying, these spam messages are sufficiently different from any of my normal e-mail that they stand out very much to the baysian algorithms.
Having run OS X server for 3 years on a mission-critical file server, I can say that Linux in the server room is quite safe. Apple suffers from a number of critical problems that currently prohibit it from taking over from Linux in the server room. First off is Apple's schizophrenia regarding the enterprise. They say on the one hand they want to be an enterprise player, but on the other hand they treat the enterprise just like the consumer market. Apple told us point blank that we cannot ever expect OS X server to not require a reboot after most updates (just like the consumer version). Hence OS X server's uptime will be at most 1-2 months, as frequent updates require frequent reboots. We complained about this to Apple, saying that Apple's OS is far far worse than Windows ever was when it comes to requiring reboots. Our rep just sighed and said that's because OS X really is a consumer product. We've also had problems with Apple's enterprise support, although that has improved somewhat.
But the most serious impedement to beating linux in the server room is the fact that although Apple ships with familiar open source products like samba, openldap, and apache, they are actually proprietary. Let me explain. So you're stuck with whatever versions apple wants to ship. You cannot upgrade or replace these OSS components with their latest versions downloaded from source. This is because apple alters and extends these programs with deeps hooks into their own OS and componentns, such as OpenDirectory. Although they release their code, it is not in the form of patches, so you cannot apply apple changes to newer versions. Further, apple-released source code is not buildable on your average OS X server machine. There are often requirements for proprietary header files (like for PasswordServer and other things including the kernel) that aren't shipped with their developer tools. I spent months trying to figure out how to build OpenLDAP 2.2 (latest at the time) on 10.3 server since their version had serious bugs. I gave up on it. Apple's official line to us in response to problems was to upgrade to Tiger server (to be fair 10.3.9 fixed many of the openldap bugs but not all).
What this really means is that Vendor locking is as bad as Microsoft and far worse than *any* other Unix or Linux vendor. At least with Solaris I can still build whatever version of Samba or OpenLDAP I want. Apple's revolutionary integration of OSS parts with nice shiny guis is seductive, but in the end, it is not done in a way that promotes the same proprietary-ness that Microsoft promotes. Our experiences with this lock-in have not been good.
Current OS X Server 10.3 uptime: 21 days (rebooted for no apparent reason -- a crash?) Current OS X Server 10.4 uptime: 1 days (just did an update) Current RHEL4 file server: 251 days (numerous updates performed, I am aware of the kernel vulnerabilities that I haven't patched yet)
Ironically we have another RHEL4 file server that I installed netatalk on to allow some mac users home directories from the that server, and AFP from the RHEL4 box is more reliable than Apple's native AFP support on 10.3 server (10.4 is better now).
Just some words of warning and caution to would-be Apple Server users.
This clip is actually very well done. The effects don't seem overly CGI wooden. They actually look good and I think would enhance the show.
I'm in favor of redoing the external shots just so long as they stay true to Roddenberry's vision and creation. I was horrified to see the USS Enterprise in Star Trek Enterprise. It wasn't anything like what Roddenberry wanted. Roddenberry always wanted his ships to be smooth with not a lot of external machinery. He felt that technology would have moved beyond the need to hack external random pipes and machines onto the exterior surface of a ship. Rather he felt all the machinery would be inside where humans could actually work on it. So while having very detailed exteriors is cool to the CG folks, it's not what Star Trek is all about.
People say things like this frequently, most often without any proof (as you have done). I think you may be right, though, but only because it's a self-fulfilling prophecy. Human laziness is an incredible force, either to impeded innovation and progress, or to cause enable it. Ironic.
As for the examples of these object-oriented, data-centric systems, by and large they are merely research curiosities, written in lisp or some other niche language. Hence they aren't studied or experimented with widely by the unwashed masses.
Re:Turning the computer inside out
on
A New Kind of OS
·
· Score: 1
Very good points. So we have two competing ideologies then. Task-based and document-based. Definitely good food for thought as the future of OSs unfold.
Slashdot Mirror
In Tiger I think it's just in the System Preferences. It's very trivial to swap Command and Option, for example, for dealing with windows keyboards. More complicated things likely require messing with the keyboard maps, which is not trivial under linux either.
The option for turning on keyboard navigation dialog boxes and stuff (that I mentioned earilier) is in the System Preferences under "Keyboard and Mouse" under "Keyboard Shortcuts." Just in case anyone is interested in that.
Will clips from shows like the Simpsons and the Family Guy start disappearing from youtube? I believe they are legal due to fair use. But we all know how copyright holders feel about that these days.
OS X is one of the few OS's to allow total customization of keyboard short-cuts. You can assign almost any combination to any menu item on any program. The downside is you have to do it from System Preferences. Although MacOS has always been mouse centric, it's actually more keyboard-friendly than windows or even linux (Gnome is only now getting good keyboard shortcut access via atk and other accessibility things). Shortcuts are very consistent and work in almost every program. Command-Q to quit, Command-W to close the window, Command-H to hide the entire app (very useful -- almost eliminates the need for multiple desktops when combined with expose and command-tab), Command-S to save, command-O to open, etc.
One thing that annoyed me to no end was the apparent lack of a way to communicate with dialog boxes using only the keyboard. Most of the time command-first letter works, but often it doesn't. I found that if I turn on some of the accessibility options in system preferences, suddenly I can tab between buttons and use the space bar to activate buttons (enter always activates the default button, not the one you're highlighting).
Knowing about how to set shortcuts, the default shortcuts, and the accessibility options has really made OS X more efficient on the keyboard for me than any other OS (well almost -- I still like activating menus on linux and windows with alt-letter). Certainly it's not as bad you illustrate.
I agree that all GUIs are lousy to a degree. Case in point is CAD software. The old autocad shortcuts (still available on autocad to this day) are the way to fly. Puck in one hand, 2 and 3 letter shortcuts in the other. Modern GUIs just don't lend themselves well to CAD.
Unfortunately I've seen a marked decrease in the effectiveness of grey-listing lately, which is similar in intent to your ideas. What I'm finding is that a lot of spam is now coming from RFC-compliant mail servers. Stock spams in particular always come through after faithfully waiting out the greylist timeout. So obviously some spammers are able to wait, even up to 45 minutes, to send their spam to me. So despite your arguments spammers will find a way to still economically spam while tolerating delays, holding connections open, etc.
From what I can tell, you're in the minority. Most users prefer to have the close button on the active tab, and many prefer to have a close button on every tab. I certainly do. I can easily close off tabs I don't want anymore without have to click on each one to raise it to the top first. Also, having a close button on each tab only really works if the tabs are always the same size. That way you can close many tabs just by clicking in the same area. Before with tab size changing this was impossible because the close button (which I had on with an extension) would always change position.
I don't mind his posts so much some of the time (I never go to his site ever), but he uses dishonest means to drive traffic to his own sites and blogs to get ad revenue. This time he says "You'll find more details and pictures about these mapping robots at ZDNet" but really it's not a ZDnet article at all; it's his blog. Slashdot editors, please either edit his posts to report the truth, or don't post his stories at all.
So why would anyone want to decide right now if they would use the GPLv3 or not? Most projects that are licensed under the GPLv2 have the "version 2 or any later version" clause in them, so developers could, when the GPLv3 is finalized, choose to "fork" their project to that license, or keep it the same.
Discussing the good and bad points of the current GPLv3 draft is valid and we should be doing that here. But to ask the question "should developers switch to it?" is immature and a little silly at this point in time. We could say "given the current draft of the GPLv3, I would not use it for reasons x,y, and z," but that is the extent of it I think.
At this moment in time, if the GPLv3 were actually released, I would probably still use the GPL v2 until I had time to really understand the v3 license, and the things it might mean to my project. Currently I think the GPLv2 has some definite weaknesses that the v3 is trying to address. For example, if I write a nice python library or module under the terms of the GPL (is that possible? I don't know offhand what license that would require), a person can just embed python in a closed-source C program, load my GPL module and use it as an integrated part of his closed-source program. The GPLv2 addresses dynamic linking from the pov of the compiler, the linker, but it doesn't take into account these other use cases that are now common. The spirit of the GPL certain forbids what I have described, unless the GPL applies to the program as well, and maybe the language does too, but it's ambiguous and would likely require a court challenge to decide. This is the type of problem the GPLv3 is trying to solve.
I hope that most GPLv2 people's objections to the v3 license will be addressed in some way. We shall see.
Yes, very good points. So the argument is about whether or not spending money in the space program brings about more economic productivity overall. I believe it can. Especially now that certain parts of the tech industry are going off-shore. Despite the consumer electronics production boom overseas (where the majority of our consumer spending seems to end up), high-tech aviation technology is still largely an American thing. Thus spending money on space and aerospace is a good thing for America, in my opinion.
With the exception of the problem of disparity and wealth distribution, everything I have said is true. Spending one dollar in the economy really does generate several times that in effect. It's similar to depositing a dollar in the bank. Due to how money flow works, one dollar in the bank increases the total money supply in an economy by three times that (depending on the reserve rate). The economy as a whole is similar to a bank, in that money is in use in several places at once. I cannot say what that increase factor is (likely it is smaller than 3 times), but logically it does exist.
Spending money on the space program would have the same effect on the economy as using fiscal policy to control inflation. Government spending in the economy does have a very real and immediate effect on the overall economy. Of course this must be done very carefully so as not to cause overheating, inflation and then later recession. As others have correctly pointed out, there is a problem of national internal debt. But if you take an extreme point of view (I do not espouse this position), paying down the debt has very little positive short-term effect on the economy, and so the debt should just be ignored (after all the money system of profits, debt, debits, credits is all really a figment of our imagination -- an immaterial house of cards).
Of course you could back up your claim with sound economic theory that refutes my statements. I would accept that.
Wasting billions of dollars? You have a strange idea of where money goes. Those billions of dollars that if spent on a space race would ultimately increase the size of our economy (national and global) by up to three fold (if I remember my Econ 101 class correctly). That amplifies the affect of the money and really allows that money to now benefit many more people, that it would if we spent it directly on, say, some kind of welfare assistance. Of course it's the disparity that we should worry about. But still. It's not like our money goes into space with the rockets.
VMWare Player. Add in a text editor and Qemu to create disk images and you've got a virtual machine. Cheap and very effective. Or just use vmware-server (free edition) to create disk images.
I highly doubt there would be any vibration at all. In order to rotate at such high speeds, the rotors need to be perfectly balanced. I don't think the noise level would be too great either. The high rpms and perfect balance would likely mean a very high-frequency noise, one you're unlikely to hear (maybe it would beach whales and make cats and dogs suicidal, though). As for fumes, I don't think you're likely to notice any smell other than the normal out-gassing your plastics and electronics already do (mmm smell that new PowerBook!).
Unfortunately battery technology in and of itself is getting close to the theoretical limit. Lithium is the most reactive metal we can use right now (unless you want to use a radioactive metal). Energy density is still slowly increasing, but this comes at the expense of longer charge times. Right now I admit Lithium technology is pretty good. Out of a 5 cell LiPoly pack, for example you can get up to 45 continuous amps for maybe 8-10 minutes. Enough to have a lot of fun at the RC flying field with a 5 pound plane. But then it's home to charge for several hours.
To answer your question, though. If the price was competitive with current battery technologies, and the energy produced as as good or greater, I'd definitely convert. In fact if the amp draw was possible, I'd stick them in my R/C airplanes. Cleaner than glow power, faster recharge cycles than electric. Oh and I wouldn't mind it in my cell phone either.
Except that we simply don't have the generating or distribution capacity to handle wide-spread electric vehicles. So we can't just use existing power lines, unfortunately. And in the short-term, one study in Canada showed that converting an entire province to electric cars would increase air pollution dramatically because of the reliance on coal-fire electrical plants (of which many more would have to be built). Now ideally, you can clean up electrical generation facilities more easily than individual cars.
Someone on slashdot once calculated the total number of megawatts we'd need to generate and deliver to power electric cars if everyone had them, and it was staggering, far beyond our current capacity (no pun intended).
DOS lives on in DOSEMU and programs like dosbox which can run fine under Windows XP: http://dosbox.sf.net/
The "Hyrdogen Economy" will never exist. It has not been and never will be an energy source. Energy storage maybe, but not a source. Even if you can take cleanly generated electricity and make hydrogen via electrolosis, I don't think it's viable, except in small applications like laptop fuel cells. Today almost all hydrogen comes from natural gas anyway. I guess these guys have just developed a more efficient way to get get the hydrogen extracted from it, but the byproducts are still the same (carbon dioxide gas).
In the long run, I think it is better to work on a carbon-neutral way of generating more complex hydrocarbons, such as bio-diesel, which will not release any greenhouse gases, yet have a very high energy density that we need. I believe it is harmful to be releasing so much net carbon dioxide into the atmosphere and the oceans.
Right. This is about software distribution, not software use by an end-user. The GPL is *not* an EULA; it's a license to distribute the program. The GPL explicitly states that you do not have to agree to the terms of the GPL in order to use the program.
I suppose that by these same arguments you could argue that EULA's must be valid because nothing else grants you the right to use copyrighted materials when you buy a program from a closed-source vendor. But EULAs go beyond copyrights into contract law, I believe they are a different beast than source code licensing.
Actually it appears that in the absence of dust storms, the color of the sky on Mars is, in fact, blue, much like earth's. In some pictures I've seen, you might mistake the martian landscape for Arizona!
http://mars-news.de/life/
and
http://mars-news.de/color/blue.html
Although take it all with a grain of salt. There are good arguments for and against the blue sky. The JPL, unfortunately has never put forth a lot of convincing proof for the red sky, fueling this debate further.
Sending rockets out into space is pretty easy, but the real trick is orbit. Cheap shots to the upper atmosphere don't do a lot of good in terms of launching satellites and other objects into orbit, although I'm sure they can provide experience with the technology. Achieving orbit requires a lot more energy. There's a reason missiles and rockets are the size they are.
Tango is just an icon set, done with SVG. They use an elipse as a shadow instead of a "real" shadow. Big deal. The icons still are a cut above XP's icon set. I dont't know much about the workings of SVG, but I'd say real shadows would be difficult but doable. In any case, it's just an icon! It's not even 3D. At 128x128 it's not going to matter that much. I think you're just being pedantic. You're welcome to submit better icon shadows to the project rather than complain about things being unprofessional. Your professional help would be greatly appreciated if you feel inclined.
Not sure why the parent comment was moderated insightful. FreeDOS has a very bright future and will continue to be used in its niche areas. So indeed it is a very useful project. So I am very grateful for the hard work of the freeDOS developers over all these years. Just the other day I used FreeDOS under dosemu to resurrect some old PowerBASIC code.
After a couple of training messages, dspam is picking these spams (with or without the animated gif) with great accuracy. Even if dspam has no idea what the spam message is saying, these spam messages are sufficiently different from any of my normal e-mail that they stand out very much to the baysian algorithms.
Having run OS X server for 3 years on a mission-critical file server, I can say that Linux in the server room is quite safe. Apple suffers from a number of critical problems that currently prohibit it from taking over from Linux in the server room. First off is Apple's schizophrenia regarding the enterprise. They say on the one hand they want to be an enterprise player, but on the other hand they treat the enterprise just like the consumer market. Apple told us point blank that we cannot ever expect OS X server to not require a reboot after most updates (just like the consumer version). Hence OS X server's uptime will be at most 1-2 months, as frequent updates require frequent reboots. We complained about this to Apple, saying that Apple's OS is far far worse than Windows ever was when it comes to requiring reboots. Our rep just sighed and said that's because OS X really is a consumer product. We've also had problems with Apple's enterprise support, although that has improved somewhat.
But the most serious impedement to beating linux in the server room is the fact that although Apple ships with familiar open source products like samba, openldap, and apache, they are actually proprietary. Let me explain. So you're stuck with whatever versions apple wants to ship. You cannot upgrade or replace these OSS components with their latest versions downloaded from source. This is because apple alters and extends these programs with deeps hooks into their own OS and componentns, such as OpenDirectory. Although they release their code, it is not in the form of patches, so you cannot apply apple changes to newer versions. Further, apple-released source code is not buildable on your average OS X server machine. There are often requirements for proprietary header files (like for PasswordServer and other things including the kernel) that aren't shipped with their developer tools. I spent months trying to figure out how to build OpenLDAP 2.2 (latest at the time) on 10.3 server since their version had serious bugs. I gave up on it. Apple's official line to us in response to problems was to upgrade to Tiger server (to be fair 10.3.9 fixed many of the openldap bugs but not all).
What this really means is that Vendor locking is as bad as Microsoft and far worse than *any* other Unix or Linux vendor. At least with Solaris I can still build whatever version of Samba or OpenLDAP I want. Apple's revolutionary integration of OSS parts with nice shiny guis is seductive, but in the end, it is not done in a way that promotes the same proprietary-ness that Microsoft promotes. Our experiences with this lock-in have not been good.
Current OS X Server 10.3 uptime: 21 days (rebooted for no apparent reason -- a crash?)
Current OS X Server 10.4 uptime: 1 days (just did an update)
Current RHEL4 file server: 251 days (numerous updates performed, I am aware of the kernel vulnerabilities that I haven't patched yet)
Ironically we have another RHEL4 file server that I installed netatalk on to allow some mac users home directories from the that server, and AFP from the RHEL4 box is more reliable than Apple's native AFP support on 10.3 server (10.4 is better now).
Just some words of warning and caution to would-be Apple Server users.
A much better version of the CGI for the same sequence can be found here:
http://www.youtube.com/watch?v=D9XHmj-dPEY
This clip is actually very well done. The effects don't seem overly CGI wooden. They actually look good and I think would enhance the show.
I'm in favor of redoing the external shots just so long as they stay true to Roddenberry's vision and creation. I was horrified to see the USS Enterprise in Star Trek Enterprise. It wasn't anything like what Roddenberry wanted. Roddenberry always wanted his ships to be smooth with not a lot of external machinery. He felt that technology would have moved beyond the need to hack external random pipes and machines onto the exterior surface of a ship. Rather he felt all the machinery would be inside where humans could actually work on it. So while having very detailed exteriors is cool to the CG folks, it's not what Star Trek is all about.
People say things like this frequently, most often without any proof (as you have done). I think you may be right, though, but only because it's a self-fulfilling prophecy. Human laziness is an incredible force, either to impeded innovation and progress, or to cause enable it. Ironic.
As for the examples of these object-oriented, data-centric systems, by and large they are merely research curiosities, written in lisp or some other niche language. Hence they aren't studied or experimented with widely by the unwashed masses.
Very good points. So we have two competing ideologies then. Task-based and document-based. Definitely good food for thought as the future of OSs unfold.